derived Neurotrophic Factor (BDNF) in Overweight ...

HMR/2012-03-0082/1.1.2013/Macmillan HMR/2012-03-0082/1.1.2013/Macmillan

Short Communication

Effect of Exercise on Circulating Levels of Brainderived Neurotrophic Factor (BDNF) in Overweight and Obese Subjects

Authors Affiliations

A. V. Araya1, X. Orellana2, D. Godoy3, L. Soto3, J. Fiedler4 1

Endocrinology Section, Hospital Clinico Universidad de Chile, Santiago, Chile Nutrition Unit, Hospital Clinico Universidad de Chile, Santiago, Chile 3 Physiatry and Rehabilitation Unit, Hospital Clinico Universidad de Chile, Santiago, Chile 4 Laboratory of Neuroplasticity and Neurogenetics, Department of Biochemistry and Molecular Biology, Faculty of Chemical and Pharmaceutical Sciences, Universidad de Chile, Santiago, Chile 2

Abstract

▼

Exercise increases the expression of brainderived neurotrophic factor (BDNF) in rodents and in healthy humans. Its relationship with weight loss and improvement in metabolic parameters, in obese human subjects, has not been elucidated. The aim of the study was to evaluate the effect of an aerobic exercise program on circulating levels of BDNF in overweight and obese subjects. We measured anthropometric and metabolic parameters in 15 male and female nondiabetic outpatients (age 38.3 ± 9.5 years, BMI 27–35 kg/m2), before and after 30 sessions of aerobic exercise (3 sessions per week). Plasma (p), serum (s), and platelet (plat) BDNF concentrations were measured at basal condition and after completing 15 and 30 sessions of exercise. Subjects were advised to continue

received 19.03.2012 accepted 12.12.2012 Bibliography DOI http://dx.doi.org/ 10.1055/s-0032-1333237 Horm Metab Res 2013; 45: 1–4 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0018-5043 Correspondence A. V. Araya Endocrinology Section Hospital Clinico Universidad de Chile Santos Dumont 999Independencia Santiago Chile Tel.: + 56/2/777 6891 Fax: + 56/2/777 6891 [email protected]

Introduction

▼

Brain-derived neurotrophic factor (BDNF), a member of neurotrophin family of growth factors, plays important role in neuronal remodeling, neuroplasticity and cognitive functions [1]. Recently, there has been increasing interest in the function of BDNF in regulation of appetite, energy expenditure, and glucose and lipid metabolism [2]. Heterozygous knock-out mice for BDNF show hyperactivity, hyperphagia, obesity, and hyperinsulinemia [3]. These phenotypes are reversed after administration of BDNF [3]. On the other hand, about 50 % of patients with the WAGR syndrome have heterozygous BDNF deletions and 100 % of them are obese at the age of 10 and have lower serum BDNF than those without BDNF deletions [4]. Further, there is strong evidence for protective role of exercise on neurons. In healthy volunteers, exercise improves cognitive function [5].

their usual food intake. A significant decrease in weight, BMI, waist circumference, diastolic blood pressure and total cholesterol was observed at the end of the study (p < 0.02). Serum and platBDNF showed a significant increase during the training period (p = 0.005 and 0.04 respectively). However, pBDNF showed no significant increase. Area under the curve of glucose at baseline, was inversely correlated with sBDNF (r = − 0.53, p = 0.04) and platBDNF (r = − 0.6, p = 0.01) after session 15. Also, platBDNF was correlated inversely with post load insulin and HOMA2IR at the end of the training program (r = − 0.53, p = 0.03 and r = − 0.52, p = 0.04, respectively). In overweight and obese subjects, serum and platBDNF levels increase after 30 sessions of aerobic exercise. This is accompanied with the improvement of anthropometric and metabolic parameters and modest weight loss.

Interestingly, upregulation of BDNF and its high affinity TrkB receptor stimulates neurogenesis in hippocampus and improves learning in animal models [6]. Recent reports have shown that aerobic exercises increase serum BDNF (sBDNF) in patients (with multiple sclerosis, depression, panic disorders) and in nontrained healthy volunteers [7]. Higher levels of sBDNF and lower plasma BDNF (pBDNF) was observed in obese subjects with type 2 diabetes than control subjects [8–10]. The reason for higher sBDNF levels in conditions associated with insulin resistance is hypothesized to be a defect in the release of BDNF from platelets. Since sBDNF is a reflection of the intraplatelet content of BDNF (platBDNF), this hypothesis explains decreased plasma levels [8, 10]. Beneficial effects of exercise on cardiovascular system and metabolism are widely demonstrated. However, there is a lack of information concerning the effect of exercise on circulating

■ Proof copy for correction only. All forms of publication, duplication or distribution prohibited under copyright law. ■ Araya AV et al. Exercise and BDNF in Obese Subjects … Horm Metab Res 2013; 45: 1–4

1

2 Short Communication


Table 1 Anthropometric and biochemical parameters of 15 obese and overweight subjects during a 3-month aerobic exercise program. Baseline n Age (years) Weight (kg) BMI (kg/m2) Waist (cm) Syst. BP (mm Hg) Diast. BP mm Hg) Total cholesterol (mg/dl) HDL (mg/dl) LDL (mg/dl) Triglycerides (mg/dl) Fasting glycemia (mg/dl) Post OGTT glycemia (mg/dl) AUCGlucose (mg/dl/min) Fasting insulin (μUI/ml) Post OGTT insulin (μUI/ml) AUCInsulin (μUI/ml/min) HOMA2-IR Plasma BDNF (ng/dl) Serum BDNF (ng/ml) Platelet BDNF (pg/platelets 106)

15 (6 M/9 F) 38.3 ± 2.4 84.4 ± 3.4 30.6 ± 0.6 108.1 ± 1.5 128.6 ± 3.7 82.3 ± 2.6

After session 30

p

83.3 ± 3.5 29.4 ± 0.6 105.5 ± 1.54 121.7 ± 3.4 73.7 ± 2

0.02 0.01 0.003 0.05 0.0007

192.7 ± 5.1 44.1 ± 3.2 117.7 ± 8 181.1 ± 23.1 96.9 ± 2.8

178 ± 7.1 45.9 ± 2.6 102.2 ± 6.7 149.5 ± 17.6 100.5 ± 2.4

0.01 NS 0.06 NS NS

132.9 ± 7.8

134 ± 6

NS

17 157.1 ± 814.8

17 140 ± 741.9

NS

13.5 ± 1.5

12.3 ± 1.2

NS

80.6 ± 9.4

71.8 ± 11.1

NS

13 437 ± 2 425.3

13 474.3 ± 3 501.8

NS

3.3 ± 0.4 0.92 ± 0.1 2.6 ± 0.45

3 ± 0.3 1.4 ± 0.2 7.6 ± 1.5

NS NS 0.005

6.65 ± 1.4

24.1 ± 5.6

0.04

AUC: area under the curve. NS: nonsignificant A p-value < 0.05 was considered statistically significant

levels of BDNF, its relationship to weight loss, and improvement in metabolic parameters, in obese human subjects. The aim of this study was to evaluate the effect of aerobic exercise on circulating levels of BDNF, weight, and metabolic parameters, in overweight and obese subjects.

Subjects and Methods

▼

Fifteen nondiabetic and sedentary subjects of both sex, between 26–51 years of age, with normal fasting glycemia, body mass index (BMI) range of 27–35 kg/m2, and without chronic diseases (such as uncontrolled hypertension, coronary disease, musculoskeletal dysfunction), alcohol or drug abuse, history of depression or other psychiatric illness, use of some medications (such as corticosteroids, antidepressants, hypnotics, oral contraceptives, or hormone replacement therapy) were recruited. Beck Depression Inventory questionnaire was used in screening to rule out undiagnosed depression. None of the subjects were treated for overweight or had received sensitizing insulin drugs in the last 2 years. This study was carried out according to the principles of the Declaration of Helsinki and was approved by the Ethics Committee of the Clinical Hospital of the University of Chile. All subjects gave an informed consent for participating in the study. Weight, height, waist circumference and blood pressure were measured in all participants. After a 10 h fasting, a blood sample

was collected for determination of glucose, insulin, lipid profile, and platelet count. For BDNF determination, blood was drawn in heparinized containers for plasma and into additive-free containers for serum BDNF. All samples were put on ice. Additivefree containers were kept on ice for 60 min to ensure complete clotting and platelet degranulation. Serum and plasma were obtained by centrifugation for 15 min at 2000 × g and 4 °C. Supernatants were obtained and stored at − 80 °C until measured. An oral load of 75 g of glucose was administrated in oral glucose tolerance test (OGTT). Glucose and insulin were measured at 30, 60, and 120 min. In premenopausal women, the sample was taken in follicular phase of the menstrual cycle. To evaluate insulin resistance, we used the homeostatic model assessment version 2 (HOMA2-IR) based on plasma fasting glucose and insulin (HOMA calculator v2.2, http://www.dtu.ox. ac.uk/). Each subject participated in an exercise program of 30 sessions, 1 h each, 3 times a week. Each session consisted of 10-min warm-up, 20 min on treadmill or bike ( ≥ 65 % VO2max), 20 min of stretching, and 10 min of relaxation. Weight, platelet count, and BDNF were measured after session 15. Anthropometric evaluation, OGTT, and lipid profile were added after session 30. Blood samples were taken 24–48 h of last exercise session. Subjects were instructed to maintain their usual caloric intake. Glucose and insulin were measured by enzymatic glucose oxidase method and automated enzyme-linked immunometric chemiluminescence assay (Immulite, Diagnostic Products Corporation, Los Angeles, CA, USA) respectively. BDNF was determined using a commercial ELISA kit (BDNF Emax® Imunoassay System, Promega, Madison, WI, USA) according to manufacturer’s instructions. Dilutions were tested in the range of 1:10– 1:100 and 1:25 dilution was chosen as it allowed detection of BDNF in all samples. The concentration of BDNF was calculated from a standard curve, which was linear between 7.8 and 500 pg/ ml. The intra- and interassay variation were 8.8 % and 18 %, respectively, and the lower detection limit was 7.8 pg/ml. All samples were processed in a single assay to minimize interassay variability. The platelet BDNF was calculated as sBDNF-pBDNF/ platelet count [8]. The total and integrated glucose response, as well as insulin responses in OGTT were calculated by the trapezoid method and expressed as the area under the concentrationtime curve (AUC) from 0 to 120 min. For statistical analysis, data were expressed as mean ± SE. BDNF did not have normal distribution. Therefore, we used nonparametric statistical tests. Friedman ANOVA and Wilcoxon rank, were used to compare differences between means before and after intervention. Pearson correlation coefficient was used to assess correlations between variables. Statistical significance was defined with a value of p < 0.05.

Results

▼

Anthropometric and metabolic parameters obtained before and ▶ Table 1. after 30 sessions of exercise program are shown in ● There was a significant decrease in weight of 1.26 kg (1.6 % of initial weight), BMI, waist circumference, diastolic blood pressure, and total cholesterol. Both, sBDNF and platBDNF increased significantly during the exercise program. sBDNF showed significant difference in levels (p = 0.005), from baseline (2.6 ± 0.45 ng/dl), after exercise sessions (4.6 ± 1.2 ng/dl after session 15 and 7.6 ± 1.5 ng/dl after session 30). The platBDNF also indicated substantial variation (p = 0.04) in levels from baseline



Short Communication

respectively). Subjects who achieved a high pBDNF level at the end of the program lost more kilograms. However, this did not reach statistical significance (r = 0.51, p = 0.05).

Discussion

▼

Fig. 1 Change in plasma a serum b and platelet BDNF levels c during a 3-month aerobic exercise program in 15 overweight and obese subjects (0 = measurement at baseline, 15 = measurement after session 15, and 30 = measurement after session 30).

(6.65 ± 1.4 pg/platelets 106 to 12.2 ± 4.4 pg/platelets 106 after session 15 and 24.1 ± 5.6 pg/platelets 106 after session 30). However, pBDNF showed no significant increase (p = 0.2), from baseline (0.92 ± 0.1 ng/dl), after sessions 15 (1.76 ± 0.5 ng/dl) and 30 ▶ Fig. 1a). We observed considerable increase (1.4 ± 0.2 ng/dl) (● in sBDNF and platBDNF in some subjects at the end of exercise program. However, others showed an increase after session 15 ▶ Fig. 1b, c). Waist and returned to baseline after session 30 (● circumference and AUCglucose were inversely correlated with sBDNF after session 15 (r = − 0.51 and − 0.53, respectively, p = 0.04). Post load glycemia and AUCglucose correlated inversely with platBDNF after session 15 (r = − 0.53, p = 0.04 and r = − 0.6, p = 0.01, respectively). platBDNF after session 15 also correlated inversely with post load insulin and HOMA2-IR at the end of the training program (r = − 0.53, p = 0.03 and r = − 0.52, p = 0.04,

It has been shown in animal models that exercise increases BDNF mRNA expression in hippocampus and some areas of cerebral cortex [11]. This explains the neuroprotective effect of BDNF associated with exercise. Many studies in humans that had shown increased sBDNF after aerobic exercise, evaluated only the acute effect of exercise [6, 12]. In our study, almost all overweight and obese subjects were observed to increase sBDNF and platBDNF, during a 3-month aerobic exercise program. An improvement in other parameters like waist circumference and blood pressure were also observed. However, modest decrease in weight and no change in insulin resistance parameters were observed. As the subjects did not modify their caloric intake during the course of study, this can be attributed to the effects of diet and caloric restriction on weight loss and also, BDNF regulation [13]. Previous studies on obese and type 2 diabetic subjects demonstrated higher sBDNF and lower pBDNF levels compared to controls. This suggests suppression of brain BDNF output by high glucose levels and insulin resistance leading to impaired release of platBDNF in such patients [8, 10]. Accordingly, we observed post load glycemia in OGTT and AUCGlucose were inversely related with platBDNF and sBDNF after session 15. Final HOMA2-IR and post load insulin were inversely related as well. These observations suggest that subjects with higher insulin resistance had lower BDNF levels and therefore lost less weight. The BDNF specific receptor TrkB is expressed in hypothalamus, pancreas, and skeletal muscle with platelets playing a possible role in the release of BDNF affecting these tissues. Remarkably all of these tissues are related to appetite and metabolic regulation. In insulin resistant subjects, exercise plausibly facilitates the release of BDNF from platelets to target tissues, affecting weight and food intake. Effects of BDNF on appetite, weight, and metabolism had been demonstrated [3, 13, 14]. Accordingly, peripheral injection of BDNF in obese or diabetic mice decreases weight and appetite, whereas improves glucose and lipid levels [14]. It is possible that in obese subjects exercise might increase tissue BDNF synthesis, increase BDNF platelet storage, and enhance BDNF release to target tissues involved in energy and metabolism homeostasis. However, BDNF is also synthesized in vascular endothelial and smooth muscle cells [15] as well as adipose tissue [16]. This explains our limitation in failing to establish the main source of BDNF build up during the training period. Another observation of our results is the variation among baseline levels of sBDNF compared with those reported in the literature. In a recent study, we have observed similar results in depressed patients (1.52–36.3 ng/ml) [17]. In a recent report, some authors showed that sBDNF levels were variable under several pathological conditions (neurodegenerative and mood disorders) and also, between control subjects [18]. Such variability can be explained by the differences in the criteria for subject recruitment, sample collection and/or manipulation. Some authors have suggested that variations in sBDNF levels could be related to mechanism in platelet peptide release, in a way unrelated to platelet reactivity [19] or, may be related to BDNF geno-


3

4 Short Communication

type variants. On the other hand, platBDNF levels are higher in women than men [8]. Thus, one limitation of our study is that we did not consider the influence of gender in our results. Nonetheless, one advantage of our study is the repeated measurements of sBDNF and platBDNF during the follow-up, validating our findings. We conclude that after 30 sessions of aerobic exercise, serum and platBDNF can increase in sedentary, nondepressed overweight and obese subjects, in spite of modest weight loss and insulin resistance improvement. Further – basic, translational, and clinical – studies are required to elucidate the role of exercise and BDNF in obese and insulin resistant subjects apart from establishing their relation to metabolic improvement and appetite regulation.

Acknowledgements

▼

This work was supported by a grant from the Bureau for the Support of Clinical Investigation (Oficina de Apoyo a la Investigación Clínica-OAIC) of the University of Chile Clinical Hospital (2006). We would like to thank to Mr. Jaime Espinoza for his technical support in samples processing, Mrs. Carolina Bienzobas for evaluating the Beck Depression Inventory, and Mr. Alvaro Reyes for his expert statistical advice.

Conflict of Interest

▼

The authors declare that they have no conflicts of interest in the authorship or publication of this contribution.


References 1 Tapia-Arancibia L, Rage F, Givalois L, Arancibia S. Front Neuroendocrinol 2004; 25: 77–107 2 Chaldakov GN. Arch Ital Biol 2011; 149: 257–263 3 Kernie SG, Liebl DJ, Parada LF. EMBO J 2000; 19: 1290–1300 4 Han JC, Liu QR, Jones MP, Levinn RL, Menzie CM, Jefferson-George KS, Adler-Wailes DC, Sanford EL, Lacbawan FL, Uhl GR, Rennert OM, Yanovski JA. N Engl J Med 2008; 359: 918–927 5 Rojas Vega S, Strüder HK, Wahrmann BV, Schmidt A, Bloch W, Hollmann W. Brain Res 2006; 112: 59–65 6 Ferris LT, Williams JS, Shen CL. Med Sci Sports Exerc 2007; 39: 728–734 7 Knaepen K, Goekint M, Heyman EM, Meeusen R. Sports Med 2010; 40: 765–801 8 Lommatzsch M, Zingler D, Schuhbaeck K, Schloetcke K, Zingler C, SchuffWerner P, Virchow JC. Neurobiol Aging 2005; 26: 115–123 9 Suwa M, Kishimoto H, Nofuji Y, Nakano H, Sasaki H, Radak Z, Kumagai S. Metab Clin Exp 2006; 55: 852–857 10 Krabbe KS, Nielsen AR, Madsen RKI, Plomgaard P, Rasmussen P, Erikstrup C, Fischer CP, Lindegaard B, Petersen AMW, Taudorf F, Secher NH, Pilegaard H, Bruunsgaard H, Pedersen BK. Diabetologia 2007; 50: 431–438 11 Soya H, Nakamura T, Deocaris CC, Kimpara A, Iimura M, Fujikawa T, Chang H, McEwen BS, Nishijima T. Biochem Biophys Res Commun 2007; 358: 961–967 12 Tang SW, Chua E, Hui T, Helmeste D, Law C. Neuroscience Letters 2008; 431: 62–65 13 Araya AV, Orellana X, Espinoza J. Endocrine 2008; 33: 300–304 14 Nakagawa T, Ono-Kishino M, Sugaru E, Yamanaka M, Taiji M, Noguchi H. Diabetes Metab Res Rev 2002; 18: 185–191 15 Nakahashi T, Fujimura H, Altar CA, Li J, Kambayashi J, Tandon NN, Sun B. FEBS Lett 2000; 470: 113–117 16 Sornelli F, Fiore M, Chaldakov GN, Aloe L. Gen Physiol Biophys 2009; 28: 179–183 17 Rojas PS, Fritsch R, Rojas RA, Jara P, Fiedler JL. Psychiatry Res 2011; 189: 239–245 18 Zuccato Ch, Marullo M, Vitali B, Tarditi A, Mariotti C, Valenza M, Lahiri N, Wild EJ, Sassone J, Ciammola A, Bachoud-Levi AC, Tabrizi SJ, Di Donato S, Cattaneo E. Plos One 2011; 6: e22966 19 Karege F, Bondolfi G, Gervasoni N, Schwald M, Aubry JM, Bertschy G. Biol Psychiatry 2005; 57: 1068–1072
